What is it like to be a PhD student at Uppsala University? In this new blog series we invite our students to tell us a little bit about their research and life as Biology PhD students.

Read on the second blog post by Biology PhD student Filip Ilievski.

The inside of a bacterial cell is a sold-out concert

By Filip Ilievski

My favorite band is Florence + the Machine. 2 years ago, I almost made it – they came in Malmö at the South Ocean Music festival. I got into the volunteering team…and I broke my arm couple of months before the concert. I didn’t make it. Florence will understand.

Figure 1. FLORENCE + the MACHINE concert. By brendenmendez from Los Feliz, USA - Florence And The Machine 26, CC BY-SA 2.0, https://commons.wikimedia.org/w/index.php?curid=68460948

Concerts are defined as events where thousands of people are packed into a venue of a restricted volume. People jumping, singing, arms waving, everyone bouncing to the rhythm of the night. And despite all that seemingly deceptive chaos, there is an order on how people are localized in a concert. The hardcore fans always try to find their way closer to the stage. They come and go in waves. The security is always stationed around the stage. They are immobile, they do not move. You and your friend go out and about – left stage wing, right stage wing, center…and you both get lost. You now need to find your friend in that chaos—not just once but follow their every move through the crowd in real time. Oh, and you and your friend are at this fully packed concert hall smaller than a grain of sand. And your movement feels like several hundred kilometers per hour when you are that small. Welcome to single-molecule tracking in living cells.

Figure 2. Single-molecule tracking – lights, camera, action! Image reproduced from Jkrieger, CC BY-SA 3.0 https://creativecommons.org/licenses/by-sa/3.0, via Wikimedia Commons.

My PhD is about developing new technologies by watching how proteins move inside E. coli bacteria. The cell interior is, absurdly crowded – way more packed than any concert you'll ever go to. Ribosomes churning out proteins, enzymes catalyzing reactions, DNA being copied and transcribed, all happening simultaneously in a space about a thousandth of a millimeter across. My job is to paint individual proteins with fluorescent dyes and then watch them dance.

The challenge in this field is pretty straightforward: to see a protein, you need to attach something glowing to it. The traditional tools – self-labeling tags like HaloTag and SNAP-tag are wonderfully bright. But they're pretty big. Attaching a 20-30 kDa tag to a protein is like, well, making your friend wear an inflatable dinosaur costume at a concert. Sure, you can spot them easily at any time, but they can't really dance anymore. Many proteins simply stop working properly when burdened with such large extra appendages.

So, it was my job to find something like a tiny glow stick – small enough that our protein could keep doing its job, bright enough that we could still track it in the cellular crowded environment.

A typical day in my PhD is all about bacteria. I take cells from agar plates and get the show going. The strains I work with aren't ordinary E. coli—they're genomically recoded organisms, engineered through years of painstaking work to speak a slightly different genetic language than their ancestors. Where normal bacteria use the genetic word "UAG" to mean "stop translating," our strains have had every single one of their 321 UAG codons swapped out. This means we can hijack UAG for our own purposes: to insert designer amino acids carrying clickable chemical handles, exactly where we want them.

Traditional fluorescence labeling is like being handed a can of spray paint and told you can only paint the very beginning or very end of each protein. My approach called FLORENCE is like being given a fine brush and permission to place a single, precise dot of color anywhere on the canvas. I prepare my cultures tubes, adding antibiotics to keep the plasmids stable, and a non-canonical amino acid called BCN-exoK that carries the chemical handle our dye will click onto. The cells don't know they're making anything unusual. Their ribosomes encounter our UAG codon and, thanks to an engineered orthogonal translation system, they obligingly insert our designer amino acid. The protein folds normally, functions normally, but now carries a tiny reactive group—a molecular Velcro waiting for its partner. Then comes the click.

Figure 3. Principle of FLORENCE labelling. Image reproduced with permission from https://pubs.rsc.org/en/content/articlelanding/2025/cb/d5cb00221d

The chemistry we use is called strain-promoted azide-alkyne cycloaddition (SPAAC), which is a mouthful that essentially describes two molecules that really, really want to react with each other. Add a dye carrying an azide group to cells containing proteins with our strained alkyne amino acid, and they find each other. No catalysts needed, no harsh conditions – just molecular recognition in the warm interior of a living cell. The dyes themselves matter enormously. Early attempts at this kind of labeling failed partly because the dyes wouldn't wash out properly. This is the same like trying to spot your glow-stick-wearing friend when everyone else at the concert has also, inexplicably, acquired the same glow sticks. We use Janelia Fluor dyes, developed at the Howard Hughes Medical Institute. These dyes are bright, stable, and critically, they wash out of cells cleanly when not attached to anything.

Figure 4. An example of FLORENCE tracking for HaloTag protein labelled with 2 different dyes. Image reproduced with permission from https://pubs.rsc.org/en/content/articlelanding/2025/cb/d5cb00221d

The imaging temperature is kept at 37°C, matching the bacteria's optimal temperature. I pipette half a microliter of cell suspension onto an agarose pad, cover it with a precision coverslip, and wait. The cells need time to settle, to resume growing. Then I start imaging.

Each fluorescent dot in my movies represents a single protein molecule. I can watch them diffuse through the cytoplasm, bind to partners, release, interact. What we've built, and what we call FLORENCE – Fluorescence Labelling in Re-coded E. coli with Non-canonical Chemical Entities. It is a complete toolkit. A re-coded bacterial strain optimized for growth, a plasmid system that efficiently produces our orthogonal translation machinery, a palette of clickable amino acids, a collection of compatible dyes.

With different click chemistries and different colored dyes, we can potentially label multiple proteins in the same cell – cellular pointillism. Red dots for one enzyme, green for its partner, watching them find each other and work together. We're not quite there yet, this paper establishes the single-color foundation, but the canvas is prepared.

I think about what questions this might let us answer. How do proteins find their interaction partners in the crowded cellular interior? How fast do molecular machines assemble and disassemble? What happens when we perturb the system with antibiotics or stress? These aren't abstract questions – they're fundamental to understanding how cells work and how they fail. And FLORENCE is one of the other existing tools to answer them! As FLORENCE is a system containing diversity, none of this would have been possible without the diverse backgrounds of all involved in the work of FLORENCE. And that is whom I acknowledge! Until the next concert!

If you like to find more about FLORENCE check out my following talks as part of the Uppsala Antibiotic Center (https://www.youtube.com/watch?v=yMSSyFr0K1M) and the GCE4all Seminar (https://www.youtube.com/watch?v=YPHsfOVzxsY).

______________________________________________________

In this first blog we hear from Matthew Bodle Cowen a fourth-year PhD student at the Department of Organismal Biology, Uppsala University. He is specializing in vertebrate palaeontology. His doctoral project explores the environments and ecosystems of the water-to-land transition.

The man who stares at fish

By Matthew Bodle Cowen

Reconstructing the distant past is like figuring out a jigsaw, except 90% of the pieces are missing, half of what you have left are ripped, and the picture can be only interpreted from faint chemical traces of the ink. Also, while you think you could find more pieces if you looked in the box, said box is up a mountain in Greenland and you don’t have funding to look for it.

I work on fish fossils from the early part of the Carboniferous, around 360 million years ago. Back then, a huge lake existed in Greenland, roughly 10 times the size of today’s lake Vänern. The lake is somewhat mysterious as, despite its vast size, we have relatively few fossils from it and they seem to represent only a handful of different taxa. How could such a vast lake have such little living in it?

We know that the lake existed during a mass extinction event known as the Hangenberg crisis. Fossils of deformed plant spores from the lake match with similar spores from other sites around the world. These spores seem to have been deformed by a burst of UV radiation, perhaps from a dying star going supernova. The fish in the lake, most commonly represented by a minnow-sized ray-finned fish called Cuneognathus, are the survivors of this event.

I’ve spent long enough looking at these specimens that they have their own names to me. 'The Rhino'.... 'Dragon-eyes'... 'Chappell'... "

A typical day for me begins with selecting specimens to work on. Each fossil has been given a label marking the year and place it was discovered, and a specimen number for reference, but I’ve spent long enough looking at these specimens that they have their own names to me. “The Rhino”, because of a protruding rostral bone. “Dragon-eyes”, because the ornamental dermal bone around the orbit looks like dragon scales. “Chappell”, because the specimen number, 2425020, puts that damn song in my head. No fish is exactly the same, but it’s striking the way they are almost all preserved in the same way. Each is on its side, flattened by the weight of the sediment that piled on top of it the second it hit the bottom. The bones of the skull are what interest me, and they too have been flattened in such a way that you can never see all of them on the same fish, but you might see all of them if you looked at enough fish. That’s my goal, to see every bone and construct a kind of platonic ideal Cuneognathus, against which I can make morphological comparisons. This will tell me what kind of fish I’m working with, and if there are any new species lurking in the dusty drawers of the collections.

Next begins the ritual of coating the specimens for photography. Beautiful as they are, the fossils are too shiny and too variable in colour to distinguish some features when seen in a photo. Hence, they must be coated to make them matte. I carry my specimens to the fume hood in the lab and begin assembling the spraying apparatus. This consists of a suggestive-looking piece of glassware containing a fine white powder that, when heated over a bunsen burner, forms a fine mist that can be sprayed onto the specimens. The powder is in fact the same ammonium chloride that coats the Saltlakrits so beloved in Sweden, and in thin layers its easy to imagine the pitch-black fossil bones are made of that confection. Exposed to too much airborne moisture, the powder will dissolve into hydrochloric acid and begin to wear my fossils just as too many godisar wears on the teeth, so I must work quickly and in a dry environment.

My samples freshly coated, I refill the tray and carry the specimens to the former “quiet office”, which houses the microscope I will need to examine them. The microscope shares the office with a post-doc and another PhD student. Both human occupants work on data obtained through high-powered synchrotron scanning, essentially a CT scan on steroids. The scans allow them to produce three-dimensional images of objects buried deep in rock, negating the need for the costly and time-devouring excavation of the sort most people think of paleontologists doing. Instead, my colleagues time is devoured segmenting- digitally picking out the fossil material from the rock in their scans. My colleagues are also working on material from Greenland, though from a different, older layer of rock to the one I work on. In their rocks they find the remains of strange, four-legged beasts- Tetrapods- the first vertebrates to set foot on the newly green land of Greenland.

I switch on the microscope and the computer attached to it, which will allow me to view my subjects on a bigger screen, and take the photos I will one day publish as evidence that I have, in fact, been doing something. Looking at the fish in front of me, I see that among the bones and scales of Cuneognathus I expect to be there are scattered pieces of another fish, an acanthodian. Ancanthodians were an ancestor of sharks and rays, but with pointy spines growing from their fins. I find these spines quite often, randomly lying in the rocks. I think of the sludgy bottom of a modern-day lake, with all bits and pieces of dead creatures lying in it. Will someone study them too, in 360 million years?

Photos taken, I clean the powder off the fossils and put them back in the drawers. I return to my own office, upload the pictures, make notes. I look through the literature on similar fish from the same era, compare with what I see in my material. Do my interpretations make sense? Did I see something new today? What should I keep in mind for tomorrow?

I know that my fossils are just a tiny part of the overall picture; a single pixel of a single piece of that vast puzzle of the past. But with each pixel revealed, the picture becomes just that little bit clearer.

-------------------------

We would love to hear from other students. Please send me a message if you are interested in writing an entry for our blog (mario.vallejo-marin@ebc.uu.se)